| commit |
1.00 |
Add new cohere2_moe model (#46115) |
|
Commit message contains explicit AI assi |
2026-05-20 |
| commit |
1.00 |
FSDP + TP & native save/load distributed (#45028) |
|
Commit message contains explicit AI assi |
2026-05-19 |
| commit |
1.00 |
Block image_start/end_token_id in generation test sampling ( |
|
Commit message contains explicit AI assi |
2026-05-15 |
| commit |
1.00 |
Fix M-RoPE device mismatch in Qwen3VL family under FSDP2 CPU |
|
Commit message contains explicit AI assi |
2026-05-15 |
| commit |
1.00 |
Extract dynamic vision/audio tensors into standalone pure fu |
|
Commit message contains explicit AI assi |
2026-05-13 |
| commit |
1.00 |
fix(testing_utils): guard get_device_capability with torch.c |
|
Explicit mention of AI collaboration: 'd |
2026-05-11 |
| pr |
1.00 |
Fix use_cache with seq_len > 1 ( #46032) |
|
PR body explicitly mentions AI collabora |
2026-05-19 |
| pr |
1.00 |
Add Sapiens2 Model |
|
PR body explicitly mentions AI collabora |
2026-05-12 |
| pr |
1.00 |
[Generation] Fix continuous batching crash when out of memor |
|
PR body explicitly mentions AI collabora |
2026-05-20 |
| pr |
1.00 |
fix: include transitive relative imports when loading from l |
|
PR body explicitly mentions AI collabora |
2026-05-18 |
| pr |
1.00 |
feat: Add support for JetBrains' `Mellum` v2 code generation |
|
PR body explicitly mentions AI collabora |
2026-05-20 |
| pr |
1.00 |
Add ctsm model |
|
PR body explicitly mentions AI collabora |
2026-04-17 |
| pr |
1.00 |
feat: Add GGUF loading support for Llama 4 (text) |
|
PR body explicitly mentions AI collabora |
2026-04-21 |
| pr |
1.00 |
fix(llama4): align MoE interface for EP/TP compatibility |
|
PR body explicitly mentions AI collabora |
2026-05-18 |
| pr |
0.40 |
add HyperClovaX Vision |
|
Polite greeting and formal intro, but so |
2026-02-27 |
| commit |
0.20 |
Warn about forgetting attention mask functions (#45811) |
|
Somewhat more formal and explanatory, bu |
2026-05-11 |
| pr |
0.20 |
Fix crash and tokenization mismatch in speculative decoding |
|
Slightly more formal, but includes techn |
2026-05-20 |
| pr |
0.20 |
Add FP8 kernel acceleration for compressed-tensors quantized |
|
Slightly formal, but domain-specific and |
2026-04-29 |
| pr |
0.20 |
Add a compatibility method so callers expecting PreTrainedMo |
|
Slightly formal phrase, but domain-speci |
2025-11-23 |
| pr |
0.15 |
Streamable chat parsing |
|
Uses informal language and domain terms; |
2026-05-08 |
| commit |
0.10 |
docs: update models architecture count and sync ACL antholog |
|
Well-structured doc change; some formal |
2026-05-18 |
| commit |
0.10 |
Fix slow Trainer path with 4D attention mask (#45852) |
|
Domain-specific detail, natural technica |
2026-05-12 |
| commit |
0.10 |
fix(rope): read original_max_position_embeddings from yarn v |
|
Technical explanation, domain-specific p |
2026-05-12 |
| pr |
0.10 |
Fix image-segmentation pipeline support for RF-DETR |
|
Concise bug explanation with domain cont |
2026-05-20 |
| pr |
0.10 |
[deepseek_v4] Add DeepseekV4NextNPredictor class for MTP dra |
|
Domain-specific detail and unfinished se |
2026-05-20 |
| pr |
0.10 |
Add SAM 3.1 to Transformers |
|
Uses technical jargon, informal summary, |
2026-05-20 |
| pr |
0.10 |
EP + Trainer integration on top of DistributedConfig (#45028 |
|
Technical explanation, context reference |
2026-05-20 |
| pr |
0.10 |
Fix _is_package_available reporting available without a vers |
|
Technical jargon and issue-focused descr |
2026-05-20 |
| pr |
0.10 |
Add video_processor support to VideoClassificationPipeline |
|
Terse, technical details with issue refe |
2026-05-20 |
| pr |
0.10 |
Add cache-aware streaming and RNN-T head for Parakeet |
|
Very technical, domain-specific abbrevia |
2026-05-20 |
| pr |
0.10 |
[loading] Fix base_model_prefix issues in conversions |
|
Informal reference to users and branch h |
2026-05-19 |
| pr |
0.10 |
Trainer.compute_loss: fix loss over-counting under TP and EP |
|
Technical context, informal language, un |
2026-05-15 |
| pr |
0.10 |
Add new cohere2_moe model |
|
Concise, informal, minimal context sugge |
2026-05-20 |
| pr |
0.10 |
🚨 [ALM] Add base model without head |
|
Motivation is concrete, uses human style |
2026-04-20 |
| pr |
0.10 |
🚨🚨🚨 [Trainer] Default to FSDP2, simplify API around fsdp + f |
|
Concise, domain-specific phrasing; no AI |
2026-04-24 |
| pr |
0.10 |
Refacto GGUF weight conversion |
|
Terse technical language, clear jargon, |
2026-03-17 |
| pr |
0.10 |
[ShieldGemma2] Support attn_implementation dispatch |
|
Technical, concise with jargon and direc |
2026-05-19 |
| pr |
0.10 |
Add Fast Image Processor for Chameleon |
|
Brief, domain-specific explanation, lack |
2025-03-31 |
| pr |
0.10 |
Fix model parallel bugs for Gemma4 |
|
Direct, technical, includes test referen |
2026-05-07 |
| pr |
0.10 |
Clean up new models after release |
|
Terse, informal explanation typical of h |
2026-05-20 |
| pr |
0.10 |
Fix models for which we don't have a dedicated tokenizer cla |
|
Informal, issue-linked, domain-specific |
2026-05-13 |
| pr |
0.10 |
:rotating_light: Modularize `ProcessorMixin` into smaller co |
|
Technical explanation, human-style modul |
2026-04-17 |
| pr |
0.10 |
skip deepgemm test except cuda |
|
Concise reasoning for skipping tests, cl |
2026-05-20 |
| pr |
0.10 |
Add MoE to Gemma4 TP plan |
|
Informal, domain jargon, brief; no AI ha |
2026-04-03 |
| pr |
0.10 |
add xpu to valid hardware for torch.compile |
|
Informal requests, domain terms, human-l |
2025-11-07 |
| pr |
0.10 |
fix test failure of speculative_generation on xpu |
|
Technical, terse, domain-specific refere |
2025-11-06 |
| pr |
0.10 |
fix some cases in xpu |
|
Brief, technical, domain references; no |
2025-11-19 |
| pr |
0.10 |
add require_deterministic_for_xpu to make the case pass in x |
|
Brief, technical, direct; matches human |
2025-11-27 |
| pr |
0.08 |
Fix wrong variable in check_model_type isinstance check |
|
Bug fix described in technical, concise |
2026-05-19 |
| pr |
0.06 |
test |
|
Bug description is technical and straigh |
2026-05-19 |
| commit |
0.05 |
Fix remaining RAG doc examples that crash on current transfo |
|
Technical, concise, domain-specific; no |
2026-05-19 |
| commit |
0.05 |
🚨 Refactor ViT to updated standards (#41693) |
|
Commit messages are terse, use domain te |
2026-05-08 |
| pr |
0.05 |
[new model] Add Zyphra/ZAYA1-8B |
|
Technical, concise, with domain referenc |
2026-05-09 |
| pr |
0.05 |
clip_grad_norm avoid O(global_params) replication |
|
Technical, refers to benchmarking and pr |
2026-05-20 |
| pr |
0.05 |
gpt_oss multi-GPU AMD support |
|
Discusses a specific hardware issue, tec |
2026-04-30 |
| pr |
0.05 |
fix(rf_detr): correct paper URL and stale checkpoint referen |
|
Minor domain-specific fix, no AI hallmar |
2026-05-07 |
| pr |
0.05 |
[CB] [Major] Rework manager to have clearer control flow + h |
|
Technical explanation, uses domain conce |
2026-05-19 |
| pr |
0.05 |
[Fix] Stranded params |
|
Problem description is technical and ref |
2026-05-20 |
| pr |
0.05 |
Fix Mamba2Mixer: use_cache with seq_len > 1 silently produce |
|
Directly references a technical bug and |
2026-05-20 |
| pr |
0.05 |
[ALM] flaky alm tests |
|
Brief, bullet style, strong domain speci |
2026-05-19 |
| pr |
0.05 |
Defensive: validate tar members in parakeet NeMo conversion |
|
Technical vulnerability reference, direc |
2026-05-20 |
| pr |
0.05 |
fix(cache): StaticLayer.get_seq_length() should return int n |
|
Technical summary, domain jargon, no AI |
2026-05-19 |
| pr |
0.05 |
Delete dead code in qwen-vl series |
|
Terse changelog style, strong domain con |
2026-05-07 |
| pr |
0.05 |
Add the other processors to auto-mappings |
|
Informal tone, concise and technical, no |
2026-05-19 |
| pr |
0.05 |
Test |
|
Brief, informal, domain-specific phrasin |
2026-05-20 |
| pr |
0.05 |
xpu output align with cuda in test case |
|
Concise, technical, human tone with doma |
2026-04-20 |
| pr |
0.05 |
fix tests/models/xcodec/test_modeling_xcodec.py::XcodecInteg |
|
Short, informal, domain jargon; typical |
2025-11-19 |
| pr |
0.05 |
extend CwmIntegrationTest to xpu |
|
Brief, informal, domain-specific; no AI |
2025-11-21 |
| pr |
0.05 |
fix tests/models/edgetam_video/test_modeling_edgetam_video.p |
|
Terse, technical, includes paste of test |
2025-12-15 |
| pr |
0.05 |
Xpu fp8 gem |
|
Direct, technical, mentions reviewers; h |
2025-12-05 |
| pr |
0.05 |
fix FastSpeech2ConformerTokenizer crash in tokenize |
|
Specific, steps for reproducing; error p |
2025-12-16 |
| pr |
0.05 |
fix failure of pytest tests/models/paddleocr_vl/test_modelin |
|
Typos, terse, references test failures; |
2025-12-22 |
| pr |
0.05 |
skip test_sdpa_can_dispatch_on_flash in tests/models/pe_audi |
|
Pastes stack traces, direct context; not |
2025-12-23 |
| pr |
0.05 |
fix TP issue within the device mesh (Tensor Parallel group) |
|
Uses domain jargon, reproducing steps, i |
2026-01-05 |
| pr |
0.05 |
fix failure of llava/pixtral |
|
Minimal, domain terms, typo; not AI-gene |
2025-12-22 |
| pr |
0.05 |
fix some case failures lead by "`torch.compile` recompiled p |
|
Casual, technical, lists cases, lacks AI |
2025-10-14 |
| pr |
0.05 |
make tests/models/sam_hq/test_modeling_sam_hq.py::SamHQModel |
|
Very terse, mentions user for review; hu |
2025-12-18 |
| pr |
0.05 |
fix crash in when running FSDP2+TP |
|
Brief, informal; includes domain-specifi |
2026-01-12 |
| pr |
0.05 |
fix `ShieldGemma2IntegrationTest::test_model` |
|
Terse with crash trace excerpt, informal |
2026-01-19 |
| pr |
0.05 |
Update `SamHQModelIntegrationTest::test_inference_mask_gener |
|
Technical, domain-specific, brief and in |
2026-01-27 |
| pr |
0.05 |
remove FbgemmFp8LinearTest |
|
Informal, includes errors and queries to |
2026-01-28 |
| pr |
0.05 |
benchmark-v2: minor fix for benchmark-v2, kernel is not corr |
|
Casual explanation with reference link, |
2026-01-23 |
| pr |
0.05 |
fix gptoss tp crash |
|
Crash stack excerpt, terse, and domain-s |
2026-02-03 |
| pr |
0.05 |
fix gptoss crash in tp |
|
Crash trace excerpt, terse, domain abbre |
2026-02-09 |
| pr |
0.05 |
fix the error of tests/quantization/fbgemm_fp8/test_fbgemm_f |
|
Brief, technical description, domain jar |
2026-01-28 |
| pr |
0.05 |
Remove accidentally added `tmp.py` |
|
Short, informal, references prior PR, cl |
2026-05-20 |
| pr |
0.05 |
fix fbgemm fp8 multi-device load failure. |
|
Brief, technical, informal tone, domain |
2026-01-29 |
| commit |
0.02 |
docs: sync legacy ACL anthology URLs and update metrics acro |
|
Commit log style, domain-specific; lacks |
2026-05-19 |
| commit |
0.00 |
[ALM] flaky alm tests (#46074) |
|
Commit messages are terse and informal, |
2026-05-20 |
| commit |
0.00 |
[loading] Free up tensors faster inside ConversionOps (#4611 |
|
Single informal word, no AI-like phrasin |
2026-05-20 |
| commit |
0.00 |
Restore test utils fix (#46065) |
|
Terse and typical bugfix summary, no AI |
2026-05-20 |
| commit |
0.00 |
Allow `ydshieh2` for now for testing migration (#46105) |
|
Short messages, no AI verbosity, co-auth |
2026-05-20 |
| commit |
0.00 |
[Fix] Remove calls to unsupported functions in TP (#46103) |
|
Uses domain abbreviations and to-do, not |
2026-05-20 |
| commit |
0.00 |
[docs] reasoning support for transformers serve (#46094) |
|
Very brief and informal, lacks any AI ha |
2026-05-20 |
| commit |
0.00 |
Fix torchao behavior for xpu (#46096) |
|
Commit messages are direct, feature huma |
2026-05-20 |
| commit |
0.00 |
🚨🚨🚨 [Trainer] Default to FSDP2, simplify API around fsdp + f |
|
Technical changelog, clear human-written |
2026-05-20 |
| commit |
0.00 |
:rotating_light: Modularize `ProcessorMixin` into smaller co |
|
Many iterations, informal notes/typos, h |
2026-05-20 |
| commit |
0.00 |
Remove accidentally added `tmp.py` (#46087) |
|
Direct, short removal message, no AI-sty |
2026-05-20 |
| commit |
0.00 |
Fix post processing RF-DETR (#46041) |
|
Short, terse commit messages; no AI sign |
2026-05-19 |
| commit |
0.00 |
[Model] Add PP-OCRv6 Models Support (#45838) |
|
Human-style init and update logs; domain |
2026-05-19 |
| commit |
0.00 |
[serve] Support for reasoning (#45690) |
|
Informal, iterative commits; no AI phras |
2026-05-19 |
| commit |
0.00 |
Fix Gemma4 generation from inputs_embeds and per_layer_input |
|
Brief, informal commit messages; human t |
2026-05-19 |
| commit |
0.00 |
Parakeet tdt (#44171) |
|
Technical jargon and detailed changelog; |
2026-05-19 |
| commit |
0.00 |
Remove mask visualization tool from `masking_utils.py` (#460 |
|
Single-word, terse commit message; human |
2026-05-19 |
| commit |
0.00 |
user friendly error when loading audio from video (#45221) |
|
Terse, informal; free-text is minimal an |
2026-05-19 |
| commit |
0.00 |
fix: owned_by field in GET /v1/models returns list instead o |
|
Concise commit summary with domain detai |
2026-05-19 |
| commit |
0.00 |
[CB] Remove OpenTelemetry (#45984) |
|
Short, domain-specific message; no AI si |
2026-05-19 |
| commit |
0.00 |
docs(readme): use canonical `huggingface.co` domain in prose |
|
Standard docs update; no AI signals in c |
2026-05-19 |
| commit |
0.00 |
Init the actual tensor, not a copy (#46030) |
|
Very terse, direct, informal—definitely |
2026-05-19 |
| commit |
0.00 |
[MultimodalLM] add language_model to the get/set_input_embed |
|
Concise, technical commit with standard |
2026-05-18 |
| commit |
0.00 |
Fix AttributeError in RAG generate() for missing config fiel |
|
Direct bugfix description using technica |
2026-05-18 |
| commit |
0.00 |
[`HRM Text`] Add integration tests (#46033) |
|
Minimal detail, informal message; typica |
2026-05-18 |
| commit |
0.00 |
Add hrm text (#46025) |
|
Casual tone, domain specifics, mixture o |
2026-05-18 |
| commit |
0.00 |
hy_v3: add XPU expectations (#45858) |
|
Single-purpose, succinct commit typical |
2026-05-18 |
| commit |
0.00 |
exaone4_5: add XPU expectations (#45890) |
|
Humans often split work in such iterativ |
2026-05-18 |
| commit |
0.00 |
hyperclovax: add XPU Expectations for CI test (#45926) |
|
Short, on-topic commit message with deve |
2026-05-18 |
| commit |
0.00 |
fix: route Granite models to TokenizersBackend to preserve t |
|
Highly technical, idiosyncratic phrasing |
2026-05-18 |
| commit |
0.00 |
fix model parallel issues for deimv2 (#45843) |
|
Multiple sign-offs, collaborative struct |
2026-05-18 |
| commit |
0.00 |
chore(ci): remove dead env vars from circleci-failure-summar |
|
Clear, domain-specific rationale with hu |
2026-05-18 |
| commit |
0.00 |
[DeepSeek V4] Fix MoE converter substring-matching FP8 scale |
|
Changelog style and commit subject are c |
2026-05-18 |
| commit |
0.00 |
[docs] adding audio/video processors (#45795) |
|
Commit log is terse, uses minimal phrase |
2026-05-18 |
| commit |
0.00 |
[CB] [Major] Add tensor paralellism (#45821) |
|
Frequent use of abbreviations, terse/cho |
2026-05-18 |
| commit |
0.00 |
bugfix(ci): avoid E2BIG in pr_slow_ci_suggestion (#45983) |
|
Highly technical language and informal, |
2026-05-16 |
| commit |
0.00 |
🚨Fix memory leaks caused by lru decorators in vision models |
|
Commit log uses terse, technical, and in |
2026-05-15 |
| commit |
0.00 |
RFDetr - use correct Roboflow org for release (#45946) |
|
Contains only summary and a human co-aut |
2026-05-15 |
| commit |
0.00 |
docs: Fix formatting issues in weightconverter.md (#45988) |
|
Brief, human-formulated commit summary. |
2026-05-15 |
| commit |
0.00 |
Fix colqwen2 test (#45981) |
|
Extremely terse, typical of human commit |
2026-05-15 |
| commit |
0.00 |
[docs] chat template prefill (#45947) |
|
Single word 'docs' is terse and human-li |
2026-05-14 |
| commit |
0.00 |
[docs] decode fast path (#45899) |
|
Terse technical commit, no AI signal. |
2026-05-14 |
| commit |
0.00 |
fix: restore `_attn_implementation `and fix request offset i |
|
Informal, includes technical shorthand a |
2026-05-14 |
| commit |
0.00 |
Support Audio Flamingo Next checkpoints (#44830) |
|
Changelog with domain jargon and informa |
2026-05-14 |
| commit |
0.00 |
Expose `per_layer_inputs` for every Gemma4 variants (#45927) |
|
Very brief commit messages; no AI indica |
2026-05-14 |
| commit |
0.00 |
chore: update benchmark_v2.yml (#45966) |
|
— |
2026-05-14 |
| commit |
0.00 |
chore: update build-ci-docker-images.yml (#45959) |
|
— |
2026-05-14 |
| commit |
0.00 |
fix(ci): set persist-credentials: false on actions/checkout |
|
Very technical, detailed, with domain-sp |
2026-05-14 |
| commit |
0.00 |
chore(ci): set default workflow permissions to contents: rea |
|
Technical explanation with human tone; n |
2026-05-14 |
| commit |
0.00 |
fix(ci): remove template injection on pull_request_target wo |
|
Technical with filenames and error refer |
2026-05-14 |
| commit |
0.00 |
chore(ci): pin all GitHub Actions and reusable workflows by |
|
Detailed and technical; lists tool and f |
2026-05-14 |
| commit |
0.00 |
[docs] ALMModelTest (#45900) |
|
Minimal commit messages; no AI signals. |
2026-05-13 |
| commit |
0.00 |
[docs] new attention mask helpers (#45949) |
|
Standard template usage and terse messag |
2026-05-13 |
| commit |
0.00 |
Enhance apply_chat_template to support custom field prefilli |
|
Contains domain-specific detail and huma |
2026-05-13 |
| commit |
0.00 |
Pass packed boundary metadata to Qwen3.5 linear-attention fa |
|
Dense technical commit log, human-like s |
2026-05-13 |
| commit |
0.00 |
BUGFIX: Support hubert models that don't have conv_pos_batch |
|
Short and focused human edits, plus coau |
2026-05-13 |
| commit |
0.00 |
Revert 45777 (#45942) |
|
Single-word revert message; human typica |
2026-05-13 |
| commit |
0.00 |
pass the otel secrets (#45933) |
|
Informal lowercase message; no AI traits |
2026-05-13 |
| commit |
0.00 |
Add initial torch_tpu backend support (#45918) |
|
Detailed technical changes and realistic |
2026-05-13 |
| commit |
0.00 |
[CB] Hide activation footprint by using the CUDA graph pool |
|
Short, informal commit language, domain |
2026-05-13 |
| commit |
0.00 |
Require input_ids for repetition penalty (#45389) |
|
Contains typos, informal style, and coau |
2026-05-13 |
| commit |
0.00 |
Fix undefined 'input' variable (#45895) |
|
Terse commit message, no AI signals, no |
2026-05-13 |
| commit |
0.00 |
Deepseek v4 csa mask collapse (#45928) |
|
Technical detail and abbreviations; no A |
2026-05-13 |
| commit |
0.00 |
[fix] Add `fatal_error` to `ContinuousBatchingManager` so th |
|
Terse, domain-specific, terse comments; |
2026-05-13 |
| commit |
0.00 |
Fix fsdp2 for gemma models with shared kv states (#45912) |
|
Casual wording and co-author trailer ind |
2026-05-13 |
| commit |
0.00 |
[docs] update model cards (#45612) |
|
Terse commit style, no AI signals. |
2026-05-12 |
| commit |
0.00 |
torch.backends.fp32_precision cascade conv/rnn so removing t |
|
Informal tone, no signs of AI generation |
2026-05-12 |
| commit |
0.00 |
:rotating_light: [`Attn`] Remove all old mask APIs from mode |
|
Changelog only, terse, notably informal. |
2026-05-12 |
| commit |
0.00 |
fix(text-generation): use token-level slicing for return_ful |
|
Terse commit summaries, human style. |
2026-05-12 |
| commit |
0.00 |
[CI] AMD docker: bump to ROCm 7.2.2 / PyTorch 2.10 + prebuil |
|
Short, task-focused fragments, human-wri |
2026-05-12 |
| commit |
0.00 |
Automatically inherit properties from children in composite |
|
One-word commit summaries, human pattern |
2026-05-12 |
| commit |
0.00 |
Fix deepseek v4 (#45892) |
|
Informal fragments, explicitly criticize |
2026-05-12 |
| commit |
0.00 |
Fix/pe audio video bugs (#45886) |
|
Concise task-based commit messages, huma |
2026-05-12 |
| commit |
0.00 |
Forward `revision` to `list_repo_files` in tokenizer loading |
|
Terse, technical commit; no AI signals. |
2026-05-12 |
| commit |
0.00 |
Remove `_checkpoint_conversion_mapping` from model attribute |
|
Brief, to-the-point message; no AI signa |
2026-05-12 |
| commit |
0.00 |
Use `_keys_to_ignore_on_load_unexpected/missing` recursively |
|
Informal, minimal changelog-style messag |
2026-05-12 |
| commit |
0.00 |
Do not keep refs to submodules internally (#45889) |
|
List of short, informal entries; no AI-g |
2026-05-12 |
| commit |
0.00 |
[docs] paper cuts (#45798) |
|
Minimal, informal phrasing with no AI ma |
2026-05-11 |
| commit |
0.00 |
[Weight converter] Revert unnecessary changes to `rename_sou |
|
Informal, pragmatic, and terse—lacks AI |
2026-05-11 |
| commit |
0.00 |
fix(minicpmv4_6): skip invalid failing tests (#45836) |
|
Terse, technical, specific—no AI style d |
2026-05-11 |
| commit |
0.00 |
audio tester class (#45391) |
|
Domain-specific shorthand and brief note |
2026-05-11 |
| commit |
0.00 |
Remove deprecation cycle for inputs embeds (#45885) |
|
Extremely brief, laconic style; no AI si |
2026-05-11 |
| commit |
0.00 |
Remove deprecation cycle for `cache_position` in masking pri |
|
Minimal, informal commit message. |
2026-05-11 |
| commit |
0.00 |
Revert "[deepseek_v4] fix CSA per-query masking in eager pat |
|
Standard revert message; no AI hallmarks |
2026-05-11 |
| commit |
0.00 |
Revert "Merge branch 'main' of github.com:huggingface/transf |
|
Auto-generated revert, template-based fo |
2026-05-11 |
| commit |
0.00 |
[deepseek_v4] fix CSA per-query masking in eager path |
|
In-depth technical detail, human tone, a |
2026-05-11 |
| commit |
0.00 |
fix(rf_detr): fix failing tests (#45845) |
|
Direct, minimal language, issue context, |
2026-05-08 |
| commit |
0.00 |
fix(granite4_vision): auto-fix failing tests (#45844) |
|
Direct technical description, signatures |
2026-05-08 |
| pr |
0.00 |
[DON"T MERGE] Forked pr to test migration - check OTEL |
|
Terse, domain-specific phrasing; no AI h |
2026-05-20 |
| pr |
0.00 |
[loading] Free up tensors faster inside ConversionOps |
|
Informal, direct; references specific PR |
2026-05-20 |
| pr |
0.00 |
[DO NOT merge!] test migration |
|
Brief, test-focused; lacks AI stylistic |
2026-05-20 |
| pr |
0.00 |
Restore test utils fix |
|
Concise, references prior PR; typical hu |
2026-05-19 |
| pr |
0.00 |
Allow `ydshieh2` for now for testing migration |
|
Casual, domain-typical phrasing; no sign |
2026-05-20 |
| pr |
0.00 |
[Fix] Remove calls to unsupported functions in TP |
|
Technical details, abbreviations; human |
2026-05-20 |
| pr |
0.00 |
Fix a regression in encoder-decoder generation cache initial |
|
Domain-specific explanation; natural, co |
2026-05-20 |
| pr |
0.00 |
Add MTP speculative decoding via MTPCandidateGenerator |
|
Technical language, domain jargon; no AI |
2026-04-24 |
| pr |
0.00 |
Qwen3 ASR and Forced Aligner |
|
Short, direct domain description; human |
2026-02-08 |
| pr |
0.00 |
add more generic support for distributed trainer tests |
|
Casual, direct ask for review; lacks AI- |
2026-05-20 |
| pr |
0.00 |
add inference mode for TP |
|
Terse, domain-specific, lacks AI-style p |
2026-05-20 |
| pr |
0.00 |
[fix] toctree |
|
Casual, minimal grammar, abbreviations, |
2026-05-20 |
| pr |
0.00 |
Fix deepspeed docker |
|
Has typos, informal tone, domain context |
2026-05-20 |
| pr |
0.00 |
Update tokenizer mappings to use TokenizersBackend for addit |
|
Direct, informal, references another PR, |
2026-05-20 |
| pr |
0.00 |
clean + fix fsdp tied weights |
|
Extremely brief, uses domain abbreviatio |
2026-05-20 |
| pr |
0.00 |
[docs] reasoning support for transformers serve |
|
Short, minimal explanation, reviewer con |
2026-05-20 |
| pr |
0.00 |
Qwen2 audio parallel |
|
Informal, uses domain jargon, reference |
2026-05-20 |
| pr |
0.00 |
Support FA2 flash_attn_with_kvcache for XPU continuous batch |
|
Technical, structured, pipeline results, |
2026-05-18 |
| pr |
0.00 |
[CTRL] Support attn_implementation="sdpa" dispatch |
|
Has technical detail and domain-specific |
2026-05-19 |
| pr |
0.00 |
Fix torchao behavior for xpu |
|
Technical, describes code and device con |
2026-05-20 |
| pr |
0.00 |
fix(quantizers): make user-supplied skip_modules additive wi |
|
Detailed, technical, no formulaic or AI |
2026-05-01 |
| pr |
0.00 |
Sequence Parallel injection as a class |
|
No free-text content; title only; clearl |
2026-05-20 |
| pr |
0.00 |
Add Molmo2 |
|
No actual free-text filled in, only temp |
2026-01-23 |
| pr |
0.00 |
fix tests/quantization/fp_quant_integration/test_fp_quant.py |
|
Terse commit message and minimal free-te |
2026-03-13 |
| pr |
0.00 |
enable tp for benchmark |
|
Brief, informal description with domain- |
2026-02-05 |
